⚛️ lattice

Inverse Problem Approach for Non-Perturbative QCD: Applications

This paper applies a novel inverse problem approach to non-perturbative QCD, successfully extracting the pion and $a_1(1260)$ decay constants and reconstructing the continuum spectral function with high precision while offering a rigorous, unified framework that eliminates the uncertainties associated with traditional duality assumptions and Borel transformations.

Original authors: Yong Zheng, Ao-Sheng Xiong, Fu-Sheng Yu, Ting Wei

Published 2026-10-30

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Yong Zheng, Ao-Sheng Xiong, Fu-Sheng Yu, Ting Wei

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The universe is built from a handful of fundamental particles, but the way they stick together to form the matter we see around us remains one of the deepest mysteries in physics. At the heart of this mystery is a force called the strong interaction, which binds quarks together to create protons, neutrons, and the pions that zip through space. While scientists have a precise mathematical description of how these particles behave when they are far apart and moving fast, the rules change completely when they are squeezed close together. In this crowded, low-energy environment, the equations become so tangled that standard calculation methods break down. This is the realm of non-perturbative quantum chromodynamics, a domain where the usual tools of physics fail to provide clear answers, leaving researchers to struggle with how to predict the properties of the particles that make up our world.

For decades, physicists have relied on a set of techniques known as QCD sum rules to peek into this hidden world. These methods work by making a specific assumption: that at a certain energy level, the messy, complex behavior of individual particles suddenly smooths out and begins to look like the behavior of the underlying quarks and gluons. This idea, called quark-hadron duality, allows scientists to bridge the gap between theory and experiment, but it requires guessing where that transition happens. It is a bit like trying to hear a single instrument in a noisy orchestra by assuming the noise stops at a certain volume; if you guess the wrong volume, your calculation of the instrument's sound will be wrong. This guesswork introduces a significant uncertainty, making it difficult to extract precise values for the properties of particles, especially for those that are excited or exist in a continuous stream of energy rather than as distinct, stable objects.

A team of researchers at Lanzhou University has now proposed a different way to solve this puzzle, one that avoids the need for guessing altogether. Instead of assuming where the transition happens, they treated the problem as a mathematical inverse. Imagine you have a machine that takes a complex, unknown shape and produces a specific shadow. If you know the rules of how light works, you can try to figure out what the original shape was by looking at the shadow. This is exactly what the researchers did with the strong force. They started with the known, high-energy behavior of quarks, which acts as the "light," and worked backward to reconstruct the unknown, low-energy shapes of the particles, which act as the "shadow." By using a rigorous mathematical technique called Tikhonov regularization, they stabilized the calculation, allowing them to extract the properties of particles without having to assume where the transition between the messy and the smooth occurs.

The team applied this new approach to the pion, the lightest particle made of quarks, and its heavier, excited cousin, the a1(1260). They also looked at the continuous spectrum of energy that exists between these distinct particles, a region that is usually very hard to study. The results were striking. The method successfully calculated the decay constant of the pion, a measure of how strongly it interacts, to be 130.8 MeV, with an uncertainty of 8.1 MeV. This value aligns perfectly with the most precise experimental measurements available, which sit at 130.57 MeV. More importantly, the researchers did not just get the number for the pion right; they also reconstructed the entire energy spectrum of the a1(1260) and the surrounding continuum. When they compared their reconstructed spectrum to real data collected from tau particle decays by the ALEPH collaboration, the match was excellent across the entire range of energies, from the lowest thresholds up to about 3 GeV squared.

One of the most significant findings of this work is that the method naturally recovers the concept of quark-hadron duality at high energies without forcing it. In the traditional approach, scientists must assume that duality holds at a certain low energy point to make the math work. In this new framework, the duality emerges automatically as the energy increases. The researchers found that above an energy of about 5 GeV squared, the reconstructed spectrum of particles smoothly merges with the theoretical prediction for quarks, showing that the two descriptions become identical. This happens without any artificial adjustments or assumptions about where the transition occurs. The method also proved that the controversial technique of Borel transformation, often used in traditional sum rules to suppress unwanted noise, is unnecessary. The results remained stable and accurate whether this transformation was used or not, suggesting that the previous reliance on it was an artifact of the older, less robust methods.

The study also provided a detailed breakdown of how different components of the theory contribute to the final results. By separating the input data into its various parts, such as the basic interactions and the effects of the vacuum itself, the team could see exactly how much each piece contributed to the mass and stability of the pion and the a1(1260). They found that the basic interactions provided the bulk of the answer, while the more complex vacuum effects played a smaller, yet essential, role. Crucially, the method allowed them to quantify the uncertainty of their results with a level of precision that was previously impossible. They could separate errors that came from the input data from errors inherent in the mathematical method itself. The analysis showed that the method itself introduced very little error, meaning that the remaining uncertainty comes primarily from the precision of the input data, which can be improved in the future.

This work represents a shift in how physicists approach the most difficult problems in particle theory. By treating the extraction of particle properties as a well-defined mathematical inverse problem, the researchers have created a tool that is both rigorous and flexible. It does not require the guesswork of defining a transition point, nor does it need the artificial suppression of certain terms to get a result. Instead, it solves for the ground state, the excited states, and the continuous spectrum all at once within a single framework. The success of this approach on the pion and the a1(1260) suggests that it could be applied to a wide range of other problems, from understanding the internal structure of protons to calculating the properties of matter under extreme conditions. For the first time, scientists have a method that can map the entire landscape of particle interactions, from the distinct peaks of individual particles to the smooth valleys of the continuum, with a clarity that was previously out of reach.

Technical Summary: Inverse Problem Approach for Non-Perturbative QCD

Problem Statement
The calculation of non-perturbative quantities in Quantum Chromodynamics (QCD) remains a fundamental challenge in particle physics. Traditional methods, such as QCD sum rules, often rely on phenomenological assumptions, including the global quark-hadron duality ansatz and the Borel transformation, to extract ground-state properties. These approaches typically require solving for the ground state first, followed by excited states one by one, leading to progressively larger uncertainties for higher excitations and the continuum spectrum. Furthermore, the reliance on an effective threshold parameter ( $s_0$ ) and the Borel parameter introduces significant systematic uncertainties that are difficult to quantify rigorously. The authors address the need for a mathematically rigorous framework capable of simultaneously reconstructing the ground state, excited states, and the continuum spectral function without arbitrary duality assumptions at low energies.

Methodology
The paper applies the "inverse problem approach," a theoretical framework previously proposed in Ref. [1], to the specific physical case of the axial-vector current correlation function. The methodology proceeds as follows:

Formulation as an Inverse Problem: The authors utilize the dispersion relation for the invariant amplitude $\Pi_q(q^2)$ of the axial-vector current. By separating the integration range at a scale $\Lambda$ (where $\Lambda$ marks the transition from non-perturbative to perturbative regimes), the dispersion relation is cast as a Fredholm integral equation of the first kind. The unknown non-perturbative spectral function $\rho(s)$ in the region $[t_{min}, \Lambda]$ is to be reconstructed using perturbative QCD and Operator Product Expansion (OPE) inputs as the known right-hand side.
Handling the Pion Pole: Recognizing that the pion pole ( $s=m_\pi^2$ ) is a known physical state, the spectral function is decomposed into an isolated pion-pole term (proportional to $f_\pi^2$ ) and a continuum term $\rho_c(s)$ . This transforms the problem from pure function reconstruction into a mixed determination of a parameter ( $f_\pi$ ) and a function ( $\rho_c$ ).
Regularization: The integral equation is proven to be ill-posed (unique but unstable). To obtain a stable solution, the authors employ Tikhonov regularization. This replaces the ill-posed operator with a family of well-posed approximations, introducing a regularization parameter $\alpha$ . The solution is found by minimizing a Tikhonov functional that balances the residual error against a penalty term enforcing smoothness on the continuum spectral function.
Prior Information and Constraints: The approach incorporates prior physical knowledge, such as the pion mass and the mass/width of the $a_1(1260)$ resonance (modeled via a relativistic Breit-Wigner distribution), to improve resolution. Boundary conditions are applied at the threshold ( $\rho_c(sth)=0$ ) and at the separation scale $\Lambda$ (matching the OPE result).
Uncertainty Quantification: A rigorous error analysis is performed, distinguishing between "input-induced uncertainties" (from OPE truncation, condensate values, and resonance parameters) and "method-intrinsic uncertainties" (from the regularization approximation and parameter variations).

Key Contributions

Simultaneous Reconstruction: The method successfully extracts the pion decay constant ( $f_\pi$ ), the $a_1(1260)$ decay constant ( $f_{a_1}$ ), and the continuum spectral function simultaneously within a single mathematical framework.
Rigorous Error Control: The paper demonstrates that the method allows for the systematic quantification of both statistical and systematic uncertainties. Crucially, it shows that the approximation error introduced by regularization is controllable and diminishes as input precision improves.
Duality without Assumption: Unlike QCD sum rules, this approach does not assume quark-hadron duality at low energies. Instead, duality emerges naturally as a consequence of the solution at high energy scales ( $s \gtrsim 5 \text{ GeV}^2$ ).
Independence from Borel Transformation: The authors demonstrate that the results are essentially identical with or without the Borel transformation, eliminating the arbitrariness associated with choosing a Borel window.

Results
The application of the inverse problem approach yields the following results:

Decay Constants: The extracted pion decay constant is $f_\pi = 130.8 \pm 8.1 \text{ MeV}$ , which agrees with the experimental value of $130.57 \pm 0.14 \text{ MeV}$ . The extracted $a_1(1260)$ decay constant is $f_{a_1} = 274.9 \pm 13.7 \text{ MeV}$ .
Spectral Function: The reconstructed spectral function reproduces the ALEPH $\tau$ -decay data from the threshold ( $s \approx 0.17 \text{ GeV}^2$ ) up to $s \approx 3.16 \text{ GeV}^2$ . The solution correctly captures the $a_1(1260)$ resonance peak and the subsequent continuum.
High-Energy Behavior: For $s \gtrsim 4 \text{ GeV}^2$ , the reconstructed spectral function smoothly approaches the OPE result, confirming the natural emergence of quark-hadron duality. The violation of duality in the range $5\text{--}10 \text{ GeV}^2$ is estimated to be only $(1\text{--}4)\%$ .
Convergence: Numerical tests show that as the theoretical input is improved (e.g., including NLO terms and condensates), the extracted values converge toward experimental data, validating the method's predictive capability.

Significance and Claims
The paper claims that the inverse problem approach offers a robust alternative to traditional non-perturbative QCD methods. Its primary significance lies in its mathematical rigor: it treats the extraction of non-perturbative quantities as a well-defined inverse problem with guaranteed convergence properties as input errors vanish.

The authors emphasize four specific advantages:

Mathematical Rigor: The use of Tikhonov regularization ensures stability and allows for the systematic improvement of precision with better inputs.
Comprehensive Error Budget: The separation of input-induced and method-intrinsic uncertainties provides a transparent and quantifiable error analysis, showing that the method itself introduces negligible bias compared to input uncertainties.
Unified Framework: It solves for the ground state, excited states, and continuum simultaneously, avoiding the progressive loss of accuracy associated with sequential extraction methods.
Natural Duality: It avoids the low-energy duality assumption and the Borel transformation, deriving high-energy duality as a result rather than an input, thereby removing associated systematic uncertainties.

The work concludes that the successful application to the pion and $a_1(1260)$ demonstrates the broad applicability of this approach to other non-perturbative quantities, such as light-cone distribution amplitudes and spectral functions relevant to the muon anomalous magnetic moment.

Technical Summary: Inverse Problem Approach for Non-Perturbative QCD

More like this